Geology Reference
In-Depth Information
specify unlikely combinations of soil and man-
agement files on these highly flexible interfaces,
such as specifying a high severity fire soil in com-
bination with a forest road management file.
Part of this problem can be addressed with
improvements to model interfaces, lookup tables
for model parameters, and internal file manage-
ment (Hall & Elliot, 2001; Flanagan
et al
., 2001).
However, as erosion models continue to become
more complex and integrate with other technolo-
gies, users will be required to have experience in
DBMS, GIS, computer operating systems, remote
sensing, Internet search engines for data gather-
ing, and graphics, as well as a good foundation in
erosion process knowledge. Few professionals
have all of the above skills.
The solution to this problem is the develop-
ment of Internet-based applications (Kingston
et al
., 2000; Elliot, 2004; Flanagan
et al
., 2004;
Kirkby
et al
., 2004; Miller
et al
., 2004). Internet-
based tools that support erosion modelling and
conservation planning include applications to
facilitate sharing datasets and software, direct
data visualization, and online simulation. Appli-
cations exist for erosion modelling at the hill-
slope, watershed and regional scales. Many of the
applications include a spatial component by
either using spatial data or displaying the results
in the form of maps representing the spatial dis-
tribution of erosion. Geospatial technology tools,
such as Internet map services, are increasingly
being used. Results from erosion models are also
components in resource or environmental plan-
ning efforts, which are increasingly using Internet-
based applications to support the planning process
(Kingston
et al
., 2000).
(e.g. server, desktop, laptop, PDA, or mobile
phone) can be time-consuming. Application of
new erosion models often requires the use of
other applications, such as DBMS, GIS, image
processing systems (IPS), and graphic software, all
of which must operate on the same OS and plat-
form. The need for different software systems
increases the computing requirements (i.e. RAM,
disk storage) of the platform, rendering some
applications inoperable on some platforms. Users
also need expertise in the different applications,
and all application data must be stored locally,
increasing storage requirements.
Client/server systems are based on generic
client/server architecture in network design
(Tsou & Buttenfield, 1998; Peng & Tsou, 2003),
referred to as 2- or 3-tier systems. A 2-tier system
is where the data and applications are located on
the same server. A 3-tier system is where the data
and applications are located on separate servers.
Client/server architecture allows distributed
clients (i.e. users) to access a server remotely by
using distributed computing techniques such as
Remote Procedure Calls (RPCs) or database con-
nectivity techniques such as Open Database
Connectivity (ODBC). All necessary applications/
models and data are hosted on the server(s).
Applications can be developed that integrate
functionality from several different software
products and models transparent to the client.
The computer resource requirements for the cli-
ent are considerably less since the client does not
need to handle data storage and management, or
to install the applications on the local platform.
In many cases the client only needs an Internet
browser and connection. However, the client-side
components are usually platform-dependent, and
each client component can access only one server
at a time (Peng & Tsou, 2003).
Distributed systems (i.e. distributed comput-
ing, distributed services) can connect to, and
interact with, multiple and heterogeneous sys-
tems and servers at the same time (n-tier systems)
and without the constraints of traditional client/
server relationships (Montgomery, 1997; Peng &
Tsou, 2003). Under a distributed architecture
there is no difference between a client- or a
17.2
Characterization of an Internet
Application
Traditional erosion models and applications are
closed centralized systems that incorporate mod-
els, interfaces and data (Miller
et al
., 2004). These
desktop systems are platform- and application-
dependent, and migrating into different operating
systems (e.g. Apple vs. Microsoft) or platforms
